Pulmonary epithelium is devided into three histopathologically distinct compartments: bronchi, bronchioli, and alveoli. Our long term goal is to characterize the morphological changes, aberrant cellular differentiation and genetic damage associated with premalignant changes and carcinogenesis in each compartment. A. Peripheral airway cell (PAC) and neuroendocrine (NE) differentiation We have shown that field cancerization in human lung is associated with alterations in the expression patterns of PAC and NE markers. We are now developing experimental models in which the pulmonary changes parallel those seen in man. Approximately one third of all human lung cancers are characterized by NE differentiation. We have previously shown that a neural transcription factor from Drosophila, achaete-schute homolog-1 is expressed at high levels in NE lung cancers and is essential for NE differentiation in neoplastic and non-neoplastic lung. We have also shown that constitutive expression of achaete-schute homolog-1 under CC10 promoter together with SV40 caused massive tumors with NE differentiation, which provides a model for a clinically important subset of human non-small cell lung cancers with NE features. Using cDNA and oligo microarray technology we have identified differentially expressed genes in this tumor model, and are currently validating their expression and studying their potential function in NE carcinogenesis. Notably, constitutive expression of the achaete-schute homolog-1 transcription factor alone under CC10 promoter also caused marked bronchialization of alveoli, which provides a model for a potential premalignant lesion for human lung adenocarcinomas. We have previously shown that such lesions occur in 12% of human lung cancer resection specimens being part of the field cancerization phenomenon. Using our mouse model we have now demonstrated that the lesions progress through proliferation and resistance to apoptosis by bronchiolar cells, rather than by cellular transformation of alveolar cells. We are currently assessing the molecular mechanisms and genetic pathways in vitro using cell lines and cDNA microarrays. B. Mouse model for human small cell lung cancer (SCLC) The prototype of human NE lung cancers is small cell lung cancer (SCLC), the most common and virulent of all NE cancers in man. One hundred percent human SCLCs contain alterations of RB1 and/or P53 tumor suppressor genes. We established a mouse model by conditional inactivation of these genes in mouse lung epithelial cells. Mice carrying conditional alleles for both RB1 and P53 developed highly aggressive lung tumors with striking similarities with human SCLC in their morphology, immunophenotype and extrapulmonary metastatic spread. In our model, inactivation of both RB1 and p53 was a prerequisite for the pathogenesis of SCLC. We are now studying the role of lung specific transcription factors and cell cycle abnormalities in the mouse SCLC model and precursor lesions. C. Generating a mouse model for squamous cell carcinogenesis in the lung. Squamous cell carcinoma composes approximately 25% of all lung cancers in the United States. It is a central tumor arising from bronchi, and the only human lung carcinoma for which the precursor lesions are well established. However, squamous cell carcinoma in mouse is rare, and there is a critical need to have a good pre-clinical model for chemoprevention studies. Our goal is to establish a mouse model for lung squamous cell carcinoma. We postulate that keratin 5/14 (K5/K14) containing cells are necessary for the formation of squamous cell carcinomas. We have shown that they are scanty or non-existent in mouse intrapulmonary airways. We have generated a novel transgenic mouse which constitutively expresses K14 in airway epithelial cells under the CC10 promoter.